Center for European Studies Working Paper No.


The Language of Democracy: Vernacular Or Esperanto? A Comparison between the Multiculturalist and Cosmopolitan Perspectives

Daniele Archibugi
Harvard University, Center for European Studies 27 Kirkland Street - Cambridge, MA 02138, USA Tel. +1(617)495-4303 Ext. 227 Fax +1(617)495-8509 Email: Italian National Research Council Via dei Taurini, 19 00185 Rome, Italy Direct Line Tel. +39-06 4993 7838 Fax +39-06 4993 7849 email

Will Kymlicka has argued “democratic politics is politics in the vernacular.” Does it imply that democratic politics is impossible in a multilingual community, whether at the local, national, regional or global level? This paper discusses this assumption and maintains that democratic politics should imply the willingness of all players to make an effort to understand each other. Democratic politics imply the willingness to overcome the barriers to mutual understanding, including the linguistic ones. Any time that there is a community of fate, a democrat should search for methods that allow deliberation according to the two key conditions of political equality and participation. If linguistic diversity is an obstacle to equality and participation, some methods should be found to overcome it, as exemplified by the Esperanto metaphor. The paper illustrates the argument with four cases of multi-linguistic political communities: a) a school in California with English-speaking and Spanish-speaking students; b) the city of Byelostok in the second half of the nineteenth century, where four different linguistic communities (Polish, Russian, German and Yiddish) coexisted. This led Markus Zamenhof to invent Esperanto; c) the linguistic problems of the Indian state, and the role played by English – a language unspoken by the majority of the Indian population in 1947 – in developing Indian democracy; and d) the case of the European Parliament, with twenty languages and a wealth of interpreters and translators.


Citizens of the world, converse! Introduction A chronicler d’exception, Gustave Flaubert, tells how, during the European riots of 1848, there were people in Paris, the city that had triggered the revolutionary rumble, who posed the problem of finding a language capable of becoming a means of communication for the new Europe: Michel-Evariste-Népomucène Vincent, a former professor, proposes that European democracy adopts a single language: a dead language might come in handy, an updated form of Latin, for example (Flaubert, 1869).1 Suggesting Latin as a new language for the continent was a way of putting all populations on the same plane and scaling down the aspirations of French as a lingua franca. A French-language dominated Europe would not have displeased the old European aristocracy (Frederick II’s and Catherine II’s French were no worse than Louis XVI’s), and it was explicitly requested during the Great Revolution. Anacharsis Cloots, for example, preached the universal republic, claiming that Paris was the capital of the world and French the planetary language (Clootes, 1793, p. 9). These wild imaginings had been hoisted on the muskets of Napoleon’s troops, tardy and incoherent offspring of the Jacobins, who had imposed universal units of measurement, codes and even festivities all across Europe – but always in French. If, in short, there were Parisians in 1848 who felt Latin ought to replace French as a lingua franca, the fact should be seen as an act of humility intended to place all nations on the same plane – including the youngest ones that, with great fervor, were only just entering the European stage. Yet the equality among nations that was to be assured by reviving a dead language did not mean rendering individuals equal too. Latin was widespread throughout Europe, but was always known by the same social classes: aristocrats, intellectuals and priests. In each country, it served to exclude the majority of the population from religious, scientific, civil and political rites. More than French, and certainly more than English or German, Latin brought together the members of the community of letters, but at the cost of excluding the great majority of the population. The idea that a common language was necessary reappeared periodically in European and world history every time revolutionary riots broke out. While many intellectuals attempted to design a “perfect” language (see Eco, 1995), the question has surfaced in politics at the Peace Congresses, in the various Workers’ Internationals and in today’s World Social Forums. Every day international organizations and the European Union have to deal with the diversity of languages, and they have to agree on new communication protocols among members of governments, bureaucrats and citizens. A partisan of democracy might point out that, though world history has seen cities, peoples and nations enter into contact by language at least as much as by the sword, the reins of that contact were held not by the majority of the population, the demos, but by an elite, the oligos. Hereditary dynasties supported by diplomats and tutors held a firmer grip on foreign policy than domestic policy, partly because they enjoyed a monopoly over linguistic communication. Democracy (demos + kratos, the power of the many) obviously seeks to break this monopoly and to do so it needs a suitable linguistic medium. This article explores the question, especially in the light of the idea that democratic politics, as Will Kymlicka (2001) has argued, can only be conveyed in vernacular languages. The article is arranged as follows: the next section discusses the nature of linguistic rights and why they have become more important in our age, while section 3 argues that, although political life can draw advantage from the existence of a sole language within a community, what differentiates democratic politics is the desire to understand not only like but also diverse individuals. In the absence of a linguistic medium, democratic politics stands out for its predisposition to create that medium, even artificially if need be, as suggested by the Esperanto metaphor. Section 4 addresses the implications of the linguistic question for our conception of democracy. In section 5, I

With detached irony, Flaubert also recounts the debate that ensued: “’No, no Latin!’ said the architect, ‘Why not?’ said the teacher. And these two gentlemen engaged in a discussion in which the others intervened too, each adding his own bit just to astonish, and it soon became so bothersome, that many went away.” 2

describe paradigmatic cases of multilingual communities and outline the policies put forward by multiculturalist and cosmopolitan theorists. In the final section, I draw some conclusions. Linguistic rights and political communities Over the last twenty years, a problem that had previously been stifled or even ignored in many countries – that of linguistic diversities within the same political community – has come forcibly to the surface. Where the languages of minorities were repressed for a long time – in Spain or the former Soviet republics, for example – the importance of diversity has reemerged, leading to multilingualism or even to secession. But new conflicts and demands have also emerged in consolidated democratic states such as Belgium and Switzerland.2 In other contexts, linguistic diversity seemed to be a consequence of changes in demographic structure: more than thirty-five million Hispanics continue to preserve their language and customs in the United States which, if present demographic trends continue, will soon be the second Spanish-speaking country in the world. For years, Berlin has been the world’s third Turkish city, while the European Union is struggling to build its own integration despite its plurality of languages. In short, multilingualism has become an indispensable part of all or almost all political communities. Outside states, international organizations – governmental and non-governmental – have increased in number and importance, and they too have to address the linguistic problem every day. Not that this is a total novelty. The Roman Empire was made up of a myriad of tribes, each with its own different language. Prior to “liberal neutrality,”3 the Romans granted each tribe ample religious and linguistic autonomy, provided they paid tribute and supplied soldiers. To preserve their empire, the Romans would take some of the most promising sons of aristocratic families hostage and provide them with education in Latin, without even asking them to pay tuition fees. The young men thus often became go-betweens for collaboration and dominion: to the Romans it was quite clear that, since they were the dominators, their language should therefore be the dominating one. After the Romans, many other communities had to come to terms with differences among languages, but it seems that these differences were tolerated insofar as individuals were subjects and not citizens. Subjects, the vast majority of whom were engaged in farming, were not expected to give voice to their thoughts, but only to work the land and pay taxes. The people were neither required nor desired to be polyglot. The resurgence of the language problem in our era is the result of two fundamental contemporary historical processes. The first has to do with the increased interdependence among distinct communities – what, to adopt a term that no one likes but cannot help using, we define as “globalization.” State political communities have become increasingly permeable to trade flows, migrations, mixed marriages and tourism. The second phenomenon has to do with the increased importance of individual rights, which has emerged both in a broadening of rights in democratic states and in an increase in the number of states in which democracy is in force. The first process is substantially guided by civil society, whereas political institutions and the pressing transformations to which they are subjected drive the second. If the state is taken as a reference point of political organization, the problem of linguistic rights can be divided into two major categories. The first regards the existence of different languages inside a state community. This is the problem the various multiculturalist theorists have at heart. The second problem concerns multilingualism in political communities outside the state or through states. In general, this is the theme that cosmopolitan theorists address. Table 1 summarizes the causes and applications of linguistic rights. Each box shows cases in which the demand for them emerges.

2 3

For a comprehensive analysis of these two cases, see McRae (1983 and 1986). By liberal neutrality I obviously refer to the idea that a liberal state should not take sides in aspects of its citizens’ private lives, including the religion they practice and the language they speak. Cf. Kymlicka (1995), Patten (2003b, pp. 365-7). If, as Kymlicka argues, the liberal state has ever promised neutrality with regard to language, this is denied by Clare Chambers (2003, p. 301). 3

Table 1 – Causes and applications of linguistic rights
Inside the state Applications: Causes: Globalization Increase in languages used inside (guided by economic and social the state on account of migration, phenomena) trade, media, cultural exchanges etc. Democratization Demand for rights for original and (guided by political-institutional newly formed linguistic minorities factors) Outside the state Growing number of communities whose destinies are transversal to the state with common problems and different languages Demand to make the international system and its organizations more transparent and accountable

Inside the state. Multilingual communities may be found in: a) multilingual states (such as Switzerland, Belgium and India); b) states with groups of immigrants who have preserved their own language (like the Hispanics of the United States); c) states that have incorporated indigenous populations who have maintained their own languages (such as aborigines in the United States and in Australia). In all three cases, the demand for linguistic rights is addressed at an existing institution that already possesses the authority, resources and explicit competences to assist its citizens. This has led states to have more than one official language (the case of Switzerland and Belgium, for example), to promote bilingualism (the case of Canada), or to allow certain regional minorities to use their own languages (the case of Catalonia, Quebec and Alto Adige). Traditionally, autocratic regimes have banned the use of the languages of minorities (suffice it to think of Italy and Spain during Fascism), in extreme cases even repressing their use in private. But as Kymlicka (1995, pp. 111 and passim) and Patten (2003b, pp. 357 and 366) have forcefully pointed out, not even liberal states have been neutral towards language. While Kymlicka argues that a liberal state cannot be neutral, Patten suggests that this is an achievable goal. The history of nation-building has traditionally involved the promotion of the official language and the repression of others, even in liberal states. States have explicitly or implicitly assumed that the linguistic minorities should accommodate to the majority language. But over the last few years, multilingualism has been the direct consequence of democratization: in Paraguay, for example, with the holding of free elections Guara nì (the idiom spoken by the majority of the population) has at last become an official language alongside Spanish. In fact, the case of Paraguay shows that the “dominant” national language has often been not the language of the majority of the population, but the language of the groups holding power. The problems to be addressed are still many and various (for a review of them, and the debate among contrasting perspectives, see Kymlicka and Patten, 2003; Castiglione and Longman, 2005). In which cases is it proper for the state to provide education in languages other than the dominant one? To what extent must the restrictions applied to education be valid for other public services such as health and social security? Ought not the right to the best defense acknowledged to anyone accused of committing a crime include the right to be tried in his or her own mother tongue? Outside the state. A growing number of problems extend beyond state political communities. Again, the linguistic dimension has acquired greater importance, both because spheres of influence outside the state have increased and because the problem of participation, transparency and accountability of public opinion in the life of existing international organizations is perceived as being increasingly important. For many years, intergovernmental organizations were considered the exclusive domain of governments: only governments participated directly in their activities, undertaking to represent the interests of their own citizens, to the extent that they could decide whether to transfer information – and what information to transfer – inside their own political communities, and this substantially reduced the linguistic problem. The job of international organizations was to facilitate communication among narrow groups – government functionaries – making available mediators such as the diplomatic corps, bureaucrats, interpreters and translators (who can be labelled linguistic intermediaries). It was then the mission of single governments to transmit important information to national public opinion. But democratization did not take place only within states: over the last fifteen years, it has also encompassed intergovernmental organizations. On the one hand, there are ever more frequent requests to make international organizations more transparent and accountable to public opinion; on the other, such organizations have begun to supply services to individuals directly, without the

intermediation of state governments. A direct relationship is thus gradually being forged between world public opinion and international organizations, and this has exacerbated the problem of linguistic communication. The transparency and control of the action of international organizations and service provision are necessarily entrusted to linguistic intermediaries. It is necessary, however, to point out a fundamental difference between demands for linguistic rights inside and above the state. As in the case of human rights, no consolidated institutions and procedures exist above the state that can assume responsibility for the respect of linguistic rights. Only in very rare cases can the individual demand rights from existing international organizations. A citizen has no direct channel to access international organizations to demand that the services are provided in her or his own language. These requests can be made by national governments only. Outside the state, the growing role of non-governmental organizations should also be considered. Amnesty International, Médecins sans Frontières and professional and cultural associations are in fact increasingly emerging as a network of global civil society.4 Internally, such organizations have often found a lingua franca to communicate with, but whenever they enter into contact with specific local communities, they too have to address a linguistic problem. The problem is by no means a new one: Karl Marx wrote the Inaugural Address of the International Workingmen’s Association in English and German; the dominant language of the Socialist International was German and this created discontent among French-speaking members; the first four congresses of the Communist International relied on a myriad of willing interpreters, obliged to make long chains of translations from one language to another, often distorting the bellicose positions of delegates. The majority of speeches at the nineteenth-century Peace Congresses were in French, but many orators resorted to consecutive translation. At recent Social Forums, often thanks to voluntary interpreters, simultaneous translation into many languages (as many as thirteen at the World Social Forum in Mumbai in January 2004) has been commonplace, even at the many dozens of workshops staged. What is the language of democracy? Charles V, a man proud to reign over a true world empire, once said: “I speak Spanish to God, French to men, Italian to women and German to my horse.” Though he was no champion of democracy, it would have been interesting to ask Charles V what he thought the language of democracy was. Though we will never hear his answer, Will Kymlicka’s has reached us loud and clear: Democratic politics is politics in the vernacular. The average citizen feels at ease only when he discusses political questions in his own language. As a general rule, only elites are fluent in more than one language and have the chance to maintain and develop their linguistic skills continuously and feel at ease discussing political questions in different languages in a multilingual atmosphere. Moreover, political communication has a large ritual component and these ritual forms of communication are characteristic of a language. Even if a person understands a foreign language in the technical sense, he may be incapable of understanding political debates, if he has no knowledge of these ritual elements. For these and other reasons, we can believe, as a general rule, that the more the political debate takes place in the vernacular, the greater the participation (Kymlicka, 2001, p. 214). If these affirmations are intended as a description of how democratic politics has evolved in the course of twenty-five centuries, it is hard to disagree: democracy developed in substantially restricted communities of people who managed to understand one another, not only through the same language but also through a set of tacit codes shared among their members. From the descriptive point of view, no one denies that a monolingual community has considerable advantages for democratic practice: all citizens (with the sole exception of those with impaired hearing) are able to take part in political life, any institution (from parliament to a local

Global civil society has become a term as popular as it is contested. For an overview of the different perspectives, see Glasius et al. (2001, 2002, 2003), Kaldor (2003), Keane (2003).


residents’ committee) can discuss and take resolutions without intermediaries, while government and all institutions can be controlled by citizens without any need for interpreters.5 But in how many political communities does this ideal situation really exist? Multicultural theorists have rightly described a real world that fails to comply with single-language or mono-ethnic states. Diversity of language and culture is a reality that is likely to increase inside each political community. I refer not only to the United States with its celebrated “melting pot” and hundreds of ethnic and linguistic minorities. Even countries such as Sweden and Finland, whose languages remained the exclusive domain of natives for centuries, have found themselves addressing new problems due to recent immigration. At the same time, problems that transcend the competences of single nationwide political communities are also bound to increase: for example, decisions on the agricultural and immigration policies of Sweden and Finland are increasingly taken in Brussels as opposed to Stockholm and Helsinki. Nor is it possible to ignore the fact that, albeit with some difficulty, democracy has managed to solve problems of linguistic communication. The United States has granted voting rights to immigrants from all over the world, and even if the President, the Congress and the Supreme Court have used exclusively English, political parties realize that, if they want to win elections, they need to attract the votes of millions of Hispanics. India too has become a state with some democratic procedures despite the diversity of its languages and a level of well-being far below that of the United States. In order to introduce democratic institutions, India had to adopt the English of its colonizers as lingua franca along with Hindi, and this proved politically less controversial than the use of Hindi only, perceived as the language of some but not all Indians (Chandholke, 2005). The same has happened in many other colonies, where the language of the colonizers has become the public language (often for restricted minorities), whereas vernaculars (often different one from another) have prevailed for private use. History has shown that language shift can be a consequence of new political boundaries as well as a change in boundaries can be a consequence of a language shift. Yet the present era poses new problems and new demands, greater than the ones experienced in the past. What can be done to address them? Neither the multiculturalist nor the cosmopolitan perspective intends to abandon the principles and values of democracy and tolerance. Despite the polemical fervor that has characterized the recent debate, the two perspectives have more points in common than is generally recognized. I wish to highlight four beliefs which, I assume, are shared by both multiculturalist and cosmopolitan theorists: 1) 2) 3) The building of nation states was an artificial process involving the creation of an “imaginary identity” in the sense clarified by Anderson (1991). In all states, the effect of ongoing cultural homogenization is to destroy local cultures and languages. Directly or indirectly, even liberal states support this process of homogenization. The diversity of the planet’s languages is a value that deserves to be preserved. Once the speed with which old languages disappear in the contemporary world has been acknowledged,6 it ought to be the job of governmental and intergovernmental institutions to preserve the linguistic variety of the planet through special cultural policies. Involving the highest number of citizens in the decision-making process is a constitutive value of democracy and it is the job of institutions to foster their participation (Beetham, 2000).


There are also important differences. Multiculturalists are keen to stress that the nation-building process leads to winners and losers and that the majority language group retains all gains. Cosmopolitans are less inclined to consider the advantages and disadvantages of the various groups since they implicitly assume that a establishing a common language provides advantages to all communities, and they tend to put aside the fact that some communities get a larger share of them. On the basis of these premises, how must democratic practice be modified to deal with the existence of multilingual political communities? To think that, in order to survive, democracy requires specific linguistic conditions is to underestimate its versatility and its capacity to evolve. It is necessary, on the contrary, to modify
Many other barriers will continue to exist (differences in ideology, in social class, in education level, in religious values, and so on). At least the technical barriers, though, that impede direct communication will not exist. Actually, a lack of language barriers will allow other social differences to emerge. 6 For paradigmatic references on the disappearance of the old languages, see Nettle and Romaine (2000) and Crystal (2000).

and extend democratic practice to enable it to live and prosper in environmental conditions – like the ones determined by multilingualism – different from those experienced to date. The fundamental difference between the multiculturalist and the cosmopolitan perspectives arguably emerges in the different answers they give to the following question: How should political communities deal with problems that cut across different linguistic communities, such as safeguarding individual liberties, maximizing participation and applying democratic procedures? Examples of “problems that cut across different linguistic communities” can be: the supply of education or health care in a multilingual neighborhood, the environmental management of a lake surrounded by two (or more) linguistically diverse states, the selection of a national or even international parliament. As far as the language problem is concerned, multiculturalism seeks to address common problems while conserving the linguistic identity of each community, thus enacting public policies which, de facto, separate communities by language. This is supposed to allow each community to conserve its own democratic procedure in the vernacular and to minimize exclusion within each community. In short, multiculturalism places the onus on the cohesion – linguistic cohesion included – of the community in question. Cosmopolitanism moves in the opposite direction. It has no intention of modifying the composition of the political community, even if, as a result of historical events, this community is made up of people who speak very different languages. In the face of common problems, cosmopolitanism seeks to apply democratic procedure, implementing public policies designed to remove linguistic barriers, even if this implies that some members of the population who are not fluent in the language used for public purposes might be somehow disadvantaged. From a normative perspective, the thesis that democratic politics has to be carried out in the vernacular is dangerous. I refer to all the political groups in North America and Europe that oppose the integration of immigrants and races, often doing so not because they are driven by authoritarian motives, but rather to preserve a high level of self-determination. Such political groups may, in good faith, think that minorities who do not speak their language could limit the democratic life of their community and that, to preserve their democracy, it is necessary to expel, isolate or naturalize those without the same knowledge of the language and even repress the use of languages other than the dominant one.7 Kymlicka’s thesis might thus lead to exactly the opposite effect to the one hoped for: instead of protecting the rights of minorities, it might even lead to their violation. For these reasons, I oppose the idea that democratic politics is politics in the vernacular with the contrasting thesis that democratic politics must be in Esperanto. I argue against the descriptive thesis whereby democratic politics is carried on in the vernacular by adopting the normative principle: democratic politics is not in Esperanto but, where necessary, it can and must be in Esperanto. Of course, I do not advocate the use of Esperanto, but, rather, the idea that it is the responsibility of individuals and governments to remove the language barriers that obstruct communication. Very regular with a limited number of words, Esperanto was one of the sources of inspiration for the Newspeak of George Orwell’s 1984. It was invented by Lejzer Ludwig Zamenhof (1889) towards the end of the nineteenth century for instrumental reasons, namely to allow communication in multilingual communities. Zamenhof grew up in the city of Byelostok, in present-day Poland, then part of Tsarist Russia, where four different languages were spoken. Not surprisingly, practical misunderstandings arose among the four communities, and Zamenhof optimistically had the idea of solving them by creating a language that each community would be able to learn easily as a second language. His ambition for this Newspeak was, obviously, much greater than that: if it worked for a small town in Eastern Europe, it might have universal value. Note that the aim of Esperanto was not to replace existing languages but to supplement them. Since then, Esperanto has attracted few but fervent acolytes in every country, but has been supplanted as an international lingua franca first by French, then by English. Other idioms – Mandarin Chinese, Hindi, Spanish, Russian – have become linguae francae in various regions of the world. Esperanto can be seen as a positive Utopia, perfectly symmetrical to the negative Utopia of Orwell’s Newspeak: whereas the ultimate aim of Newspeak was to repress thoughts against authority, the aim of Esperanto is to facilitate communication between indiThis is the case of the “English Only” movement in the United States, cf. Crawford (2000). More recently, similar concerns have been expressed in the provocative article by Huntington (2004).

viduals in the remote areas of the world, just as the introduction of universal weights and measures has sought to make economic and social life transparent by breaking down informational asymmetries among individuals and social classes. When a linguistic medium is lacking, the prerequisite for institutions and individuals to take part in democratic life is to create one – if need be artificially. The universal language is thus the key to cosmopolitan citizenship. What is democratic politics? The issue of language diversity within the same political community brings to the fore many aspects of the conceptions we have of democracy. If we espouse the “aggregative” model – the conception of democracy that favors the aggregation of preferences (as opposed to their formation) – the problem of language is considerably reduced. The single members of the political community (electors) already have a definite set of choices before them, and if the political community is made up of individuals speaking different languages, it is sufficient and technically possible to make the various options available in the various languages. In an aggregative model of democracy, a political community would be able to run elections easily by providing information in all the necessary languages. It ought to be the duty and interest of each political party to make its program accessible to voters in the most appropriate linguistic medium. In this model, electors are expected to formulate their preferences and check that the political party that has won the elections carries out its program, while their direct participation in political life is reduced to a minimum. If citizens were granted access to the administration and public services, language problems would obviously arise, but it is not impossible, as required by advocates of multiculturalism, to provide public services such as education and health in the languages most spoken by citizens. In many regions where two linguistic communities live side by side, public functionaries are already bilingual and the basic public services provided in more than one language. But both multiculturalists and cosmopolitans might hold the view that the aggregative model is not an accurate description of how democracies effectively operate, still less of how they should operate. Both approaches might favor a different model of democracy, which has been defined as deliberative by Habermas (1998), discursive by Dryzek (2000), and communicative by Young (2000). In this model, the essence of democracy is to be found in communication, that is to say in the capacity to understand the reasons of others and to expound one’s own. In many respects, the two models of aggregative and deliberative democracy are not opposed (as they are too often believed to be), but, rather, two phases of the same process. The first phase is that of the formation of parties and political programs, in which dialogue and persuasion prevail. The second phase is that of choices and the aggregation of preferences at election time, during which the competitive arguments of political parties prevail. If deliberation is an important part of the democratic process, the language issue becomes crucial. Patten (2003b, p. 379) argues that “a common public language is not necessary for deliberative democracy.” This position is antithetical to the one of Kymlicka who, as we saw above, even goes so far as to argue that democratic politics can only be in the vernacular. Patten imagines the technical possibility of using translators and interpreters. In a national legislative assembly, it is certainly possible to make use of simultaneous translations (examples already exist in this respect), but the more the level of politics narrows down, the more the possibility of resorting to linguistic intermediaries decreases. A “strong” democracy (Barber, 1984) stands out partly for its more diffused, less formalized procedures: local residents’ committees, parent-teacherstudent associations, parties, and trade unions are all vital components of political life. If, in short, we abandon the merely aggregative conception of democracy, the variety of languages emerges as a major practical hurdle. Yet I do not believe it is possible to generate democratic culture, if the single components (be they neighborhoods, schools, grassroots associations, parties, trade unions or local government) are not prepared to accept the principle of the inclusion of participants, irrespective of their linguistic capability. It would certainly infringe on all principles of democracy, if the various groups were to be defined on the basis of religious, economic or cultural criteria. So why should we consider the creation of linguistic confines as less atrocious? Where an obstacle to participation exists, it is up to democratic politics to seek to remove it.


To ask citizens to make an effort to understand each other is not a neutral act with respect to the conception of democracy preferred, and language is just the most evident side of mutual understanding. Understanding others requires patience and an investment of time and resources in education that might prove useless outside the political sphere. One example that comes to mind is that of the (few) Berliners who have learned the rudiments of Turkish to communicate with an essential part of their city’s population. To ask citizens to make this effort is to opt for the freedom of the ancients as opposed to that of the moderns (to use Benjamin Constant’s terminology), in so far as it means asking members of the community to devote time and energy to overcoming existing barriers to communication, even if this serves only for democratic practice. A cultural cosmopolitan is inclined to see an intrinsic and not only instrumental value in the opportunity of knowing an extra language. Policy options: a comparison between multiculturalists and cosmopolitans Arguably the best way of understanding the differences between the multiculturalist and cosmopolitan positions is to address specific cases. In this section, I discuss four paradigmatic cases: a neighborhood school, a multilingual city, a great multi-ethnic country and a supranational parliament. We obviously find significant differences both among multiculturalist theorists8 and among cosmopolitans (in particular, between ethical and institutional cosmopolitans). Though I do not attempt to represent all the various positions faithfully, I do believe that it is useful to outline paradigmatic cases – if need be forcing them somewhat – to identify the differences between the two approaches. A state school in California. In a state school in a district of Pasadena, California, traditionally dominated by English-speaking pupils, demographic trends and waves of immigration are causing a sizeable growth in the number of Hispanic pupils. Since a certain demographic decline has been recorded among the Anglos, the school manages to assimilate the new Hispanic students quite easily; indeed their presence has saved the school from being closed due to lack of numbers. The problem is that the two communities differ in terms of income level, culture, religion and language. The overall demographic data for Pasadena show that English is spoken at home by 55 percent of the population only, and that nearly 30 percent speak Spanish. These facts are obviously reflected in schools: the Hispanic students do not speak English well and their parents speak it even worse. School parent-student meetings end in pandemonium with the Anglos complaining that their children are starting to make a lot of spelling mistakes and the Hispanics protesting because their children are being bullied. At the end of a stormy meeting, an Anglo father, citing Samuel Huntington, invites the Hispanic community to dream in English. In return, an outraged Mexican slaps him in the face!9 The headmaster, a man with a fine sense of intuition, perceives that the Anglos are worried that the identity of their neighborhood is going to be lost. The Hispanics also have identity problems, and they are worried because their children receive lower marks than the others. Hispanics aren’t even as good at sports as the Anglos, largely because the principal game played is American football. A number of the Hispanic parents were born and bred in the United States, but they still do not have a great command of English. Since many of them are cleaners in the homes of the Anglos, their aspiration is to enable their children to live in conditions that will avoid perpetuating the class division based on different ethnic groups. The headmaster calls in a multiculturalist researcher and asks him to study the problem and come up with a solution. After a few weeks, the researcher submits a prospectus in which he divides the pupils into different sections, A and H. Adopting an ingenious restructuring program, he demonstrates that it is possible to teach in English in section A and Spanish in section H. The parents are free to choose the section they want for their children, though Anglos might be expected to enroll theirs in A and Hispanics to enroll theirs in H. Without any extra costs, the project also envisages the teaching of the other language in both sections, allowing the Anglos to pick up some Spanish and Hispanics to study English as a second language. The multiculturalist also notes that sport is a central element of group identity and that it would be wrong to prevent

In particular, between Kymlicka (1995) and Parekh (2002). For a variety of perspectives on multiculturalism, see Kelly (2002). 9 Huntington (2004, p. 45): “There is only the American dream created by an Anglo-Protestant Society. MexicanAmericans will share in that dream and in that society only if they dream in English.”


Hispanics from playing the game they prefer and perform best at. Hence, while American football is to be played in section A, soccer will be introduced to section H. The headmaster is puzzled. He wonders whether the project complies with the American constitution, and though California has been granted constitutional derogations, he decides to call in a cosmopolitan researcher for a second opinion. A few days later, the cosmopolitan submits his project. On the frontispiece is a quotation from Thomas Pogge (2003, p. 118): “ … the best education for children is the education which is best for each child.” The plan envisages that all pupils receive the same education in English, since this is the dominant language in the country in which they live and also the dominant lingua franca worldwide. It includes tables which show that American citizens with a good knowledge of English have: a) higher incomes, b) less risk of being unemployed, c) less risk of being imprisoned, d) better hopes for a longer life. Another table shows how English is snowballing as a second language in every continent, and asks whether it is the public school’s job – at least in terms of statistical probability – to condemn the pupil to earn less, risk being unemployed, end up in jail and even live shorter lives in order to preserve the language of his or her linguistic community. As regards sports, the study proposes the adoption of baseball, popular in both the Caribbean and in North America. Not content with demonstrating once and for all the advantages for the well-being of young pupils of teaching in English, the cosmopolitan also suggests introducing compulsory courses of Spanish language and culture for all, proposing as core subjects for a common identity the myth of Zorro, Ernest Hemingway and Isabel Allende. The adoption of a single section makes it possible to save money, which the researcher suggests using for evening courses in English for the parents of the Hispanics. Preempting a predictable objection from the Anglos – namely that the parents of the other ethnic group will accumulate more resources – the cosmopolitan proposes evening courses in salsa and other Latin American dances for the Anglo parents. He also proposes setting up a tourist association to organize holidays in the Caribbean and Central America. After reading the project carefully, the headmaster is still perplexed. The Byelostok problem. An emblematic case is that of Zamenhof’s home town of Byelostok. As mentioned above, in the second half of the nineteenth century four linguistic communities lived in the town: Poles (3,000), Russians (4,000), Germans (5,000) and Jews (18,000). This created many practical problems for commerce, education and basic public life that the Tsarist regime permitted in a territory that it had conquered only relatively recently. The most populous linguistic community, the Jews, did not have a large written corpus to rely upon in their own vernacular language, Yiddish, whereas two other linguistic communities, the Germans and the Russians, could count upon the consolidation of the language and culture of the two great bordering states, Germany and Russia. Acknowledging the difference, a multiculturalist would probably have suggested setting up four ethnic councils, each endowed with broad autonomy over the provision of services such as education and health. He or she would also have set up a “Chamber of Compensation” to help citizens exchange their homes if they wished, to make the city divisible into four homogeneous linguistic neighborhoods. This would have greatly reduced problems of linguistic misunderstanding in commerce and facilitated education in the languages of the four communities. As we have seen, the ingenious solution of Zamenhof, a true champion of cosmopolitanism, was to create an artificial new language, Esperanto, designed to place the various communities on the same plane and, moreover, to enable them to communicate with all the citizens of the world. The fact that the solution was unworkable should not prevent us from admiring its grandiose ambition, whereby a local problem was intended to provide the thrust for a universal language. A less ingenious solution – but arguably likely to yield more tangible fruits – would have been to create bilingualism for education and public communication in the main Slav language (Russian) and German (which bears many resemblances to Yiddish), allowing and developing the private use of other vernacular languages. Though constituting the absolute majority in the city, the Jewish community might have been concerned by this solution, but it would have taken into account the fact almost all the members of the community had a certain fluency in at least one other language. Zamenhof would probably have agreed with Van Parijs’s proposal (Van Parijs, 2003, p. 167), according to which the linguistic communities required to study the language of the others, in this case the Jews and


Poles, would be entitled to tangible compensations from the communities not required to study other languages.10 The case of India. India, unique for its infinity of ethnic groups and languages, is the second largest population in the world and totals about a sixth of the world’s inhabitants. Nonetheless, following independence, India managed to constitute a parliamentary democracy that has been relatively successful for a developing country (see Kohli, 2000). This has been possible, thanks in part to a national Parliament, whose members are elected in all the federal states. The best approach to the linguistic problem has proved to be pragmatism, accompanied by a healthy dose of flexibility and tolerance. Unlike in Italy, all attempts to create a unitary language as a means of fortifying national identity have so far failed (see Annamalai, 2001). The desire to create an Indian identity on the basis of a common language, Hindi, different from that of the old English colonizers – supported by none other than Mohandas Gandhi, among others – has proved to be a factor of division rather than of union. To solve linguistic conflicts, it has thus been established that communication between central government and the single states may be made in both Hindi and English. The country currently boasts as many as eighteen official languages, very few compared to the 1,650 languages actually spoken. A system has thus been created in which vernacular languages are used locally, one of the official languages is used for the political life of the single states, and the languages of communication for national politics are, de facto, Hindi and English (Chandholke, 2005, p. 44). A multiculturalist would notice immediately that Indian democracy is limited by the fact that the members of linguistic minorities have no possibility of controlling the acts of Parliament and Government. In the Parliament itself, the variety of different idioms means that there is no certainty that the members of linguistic minorities are able to understand each other. For a multiculturalist, it might have been fitter if, in 1947, India had been separated into twenty independent states instead of just two. This would have allowed each community greater political participation in their vernacular languages, and even though none of the twenty independent states would have been linguistically homogeneous, it would have been possible to protect linguistic minorities by adopting the policies multiculturalists champion in countries such as Canada or Spain. A cosmopolitan, on the contrary, would see the formation of a great nation in the wake of British colonization as a great advantage for the populations of the geographical area in question. In all likelihood, the formation of a federal state was in fact the best form of protection for the various ethnic, religious and linguistic minorities. Without it, conflicts would have broken out in the Indian peninsula as bloody as the ones that took place in the course of the subdivision of the Indian Union and Pakistan in 1947, nor can we rule out that interstate conflicts would have been generated analogous to those that have dominated African political life over the last sixty years. The fact that everyone can consider themselves Indian irrespective of language has reduced political violence, and the fact that each individual has been allowed to speak his or her own vernacular language has prevented traumatic changes of identity. Though Indians did not choose their colonizers, the fact that they spoke English rather than Dutch or Portuguese gave India a notable advantage in so far as the country has thus had direct access to the dominant contemporary idiom. True, this has so far favored elites as opposed to the majority of the population,11 but today suitable education policies can make English a notable competitive advantage for the development of Indian society. Looking to the future, a multiculturalist would probably seek to increase the number of official languages, along with local political autonomy and the preservation and teaching of the various vernacular languages. This would lead to a greater conservation of local languages but also more difficult economic, social and political integration at both national and international levels. A cosmopolitan, on the contrary, would tend to invest more in education in English alongside local languages to make English the intra-national and international lingua franca. The consequences would be the opposite of those wished for by multiculturalists: many of the local languages would probably be lost, but India would gain in terms of both national and international integration.
Van Parijs’s proposal could be implemented at least in the academic community, where English has asserted itself unequivocally as a lingua franca, and where the most diffused, read and cited academic journals are Anglo-American. This offers English native speakers a notable advantage and all the rest a notable disadvantage. By way of compensation, it would not be a bad idea for academics from other countries to ask their privileged native English-speaking colleagues to correct their howlers. 11 Estimates provided by the International Corpus of English indicates that the portion of the Indian population able to speak English vary from 4 to 20 percent (

The European Parliament. The European Parliament currently has twenty official languages. So far the number has increased together with that of the member states of the European Union. De facto, the official languages are those of the member states. There are no official languages for sub-state linguistic communities (the most significant claim for recognition being that of Catalan). The members of Parliament rely on simultaneous translation, and documents are translated into the official languages. As official languages have increased, so the translation procedure has grown more complex: there are currently 20 x 19 = 380 possible language communications (“into” and “from”), and finding interpreters capable of translating, for example, from Portuguese to Slovak or Lithuanian to Maltese and vice versa is often impossible. Hence the recourse to “double translations” (for example, from Portuguese to French and from French to Slovak). Yet even this vast linguistic “menu” fails to accommodate all the European languages and – albeit rarely – members of linguistic minorities sometimes speak their own mother tongues. The problem has become greater with the expansion of member states and it will continue to grow when Bulgaria, Rumania and possibly Turkey join the EU. Of the close on 5,000 employees of the European Parliament, 340 are translators and 238 interpreters, but the multiplication of languages could lead to a doubling of this figure. In a situation of this kind, the problem understandably arises of reducing the number of official languages of the European Parliament, though, politically speaking, it is a thorny one (see Mamadouh, 2002; Phillipson, 2003; Van Parijs, 2005). The advantage would be more effective debating, whereas the disadvantage would be — de facto if not de iure — the limitation of the passive electorate to elites who speak foreign languages. The members of the European Parliament may express themselves in any official language (Art. 117 of the Rules of Procedure of the European Parliament), though they generally speak in the language of their country. Willy Brandt was one of the first members to make a speech to the Parliament in a language that was not his mother tongue, speaking in English as opposed to German.12 He opted for self-translation. His choice was justified by the fact that the number of members who understood English was far higher than that of those who speak German. The choice was welcomed with warm applause and a few whistles. Multiculturalists would probably have whistled at him since he would have been incomprehensible for electors in his own constituency, who are nevertheless entitled to exercise control over their elected member. Brandt also compelled his German colleagues who did not understand English (possibly because they were not members of an elite) to listen to the speech of a fellow countryman in translation. Cosmopolitans would have applauded him warmly in so far as he was reducing the linguistic distance between members of Parliament, hence promoting a common language for European politics. Today proposals are being made to reduce the official languages to two, three or four, and the organs of the European Parliament have also posed themselves the problem of limiting the extensive use of interpreters and translations. Multiculturalists are probably hostile to these proposals because they would reduce the number of effectively eligible candidates (only citizens with a good knowledge of at least one official language could perform their role as members). Furthermore, though all parliamentary documents would continue to be available in the twenty or more official languages, there would always be the danger that an Assembly working in only a few languages would distance itself from the electorate and ultimately turn into an oligarchy. The cosmopolitans, on the contrary, believe that communication in one or a few languages would make parliamentary debate more authentic and direct (Van Parijs, 2005). They would suggest leaving just two official languages, English and French, and placing all members on the same plane, asking the English to speak in French and the French to speak in English. They would also point that, albeit elected in one country, Euro MPs have to respond to the population of Europe, not only to their own constituency. Besides, to be able to work well in a legislative assembly, it is necessary to be able to speak, informally if need be, with one’s colleagues, and to do this it is necessary to have knowledge of the most common languages. To avoid being esFor precision’s sake, it is worth adding that, perhaps ignorant of the proposal made 131 years earlier by MichelEvariste-Népomucène Vincent, the extreme left-wing Italian member Mario Capanna provocatively made a speech in Latin in the session of November 13, 1979, spreading panic among the interpreters’ booths. One of the few to understand his speech perfectly was the Euro MP Otto of Hapsburg, a direct descendant of the Royal House of the AustroHungarian Empire, and an elected member for the right-wing Catholic Party (CSU) in Bavaria. His family lost Lombardy-Veneto in 1860s and his Italian was rusty, so he congratulated his colleague in Latin. This was perhaps one of the last occasions in which, albeit at different ends of the political spectrum, European elites communicated in Latin.

corted by a squad of interpreters, each member of the Parliament should be able to communicate with colleagues in a common language. In short, the cosmopolitans would prefer an impoverished but directly understandable language to a myriad of more colorful yet non-accessible languages. A parliament in which each member speaks a language incomprehensible to others is not only ridiculous, but also useless. For linguistic cosmopolitanism The cosmopolitan position is founded on an assumption that needs to be made clear: namely, that nothing prevents human beings from mastering two or more languages.13 Recent linguistic research clearly demonstrates that there is no obstacle to children learning two languages,14 and whole countries in the civilized world implement compulsory education programs to enable students to learn properly not only their own mother tongue but also English. This is not necessarily to the detriment of the vernacular language, whose cultural value may be better understood (as an expression of the variety of humankind) precisely by individuals who speak more than one language. Polyglots are capable of appreciating the value of linguistic diversity much better than the illiterate. To master a universal language is not to relinquish the language of one’s own ethnic group. A more realistic solution than Esperanto was suggested by Aldous Huxley in The Island (1962), a novel that describes an utopian small community in the Pacific, the imaginary island of Pala, as advanced as it is rooted in its own traditions. This community preserves its own local language, but all of its members speak English, and this allows them access to technology, information and culture from the most advanced regions of the world. In the real world, the countries with the highest indices of human development – Norway, Sweden, Holland – are very close to Huxley’s ideal. On a planet on which one-third of the population is still illiterate it is undoubtedly innovative to think in terms of institutionalizing a sort of bilingualism. It comes as a surprise to find out that two-thirds of the planet’s inhabitants are already bilingual today (Baker and Prys Jones, 1998, section 1), but this still fails to bring the peoples of the world together, simply because no language of communication exists: in a word, what is lacking is a common language spoken by everyone as a second or third language. Yet, in the course of two or three generations, it may be possible to find a widely diffused linguistic medium. Rather than choose today between Vernacular and Esperanto, it might be more useful to support investment in education to allow individuals to increase their language skills. In India and Europe, multilingualism can already be seen in action (see Laitin, 1997; Mamadouh, 2002; Van Parijs, 2005; for Europe, and Annamalai, 2001; and Chandholke, 2005; for India). The British in Europe and Hindis in India are among the privileged who can afford to speak a single language, whereas many others have to speak at least two (English as lingua franca and their own vernacular language), others still already speak three (like the Catalans who need to speak Spanish as the dominant language of their state, and English as the dominant European and international language). I do not intend to argue that linguistic access is open to all: as Kymlicka rightly points out, elites are still at an advantage and, in a globalized world, also enjoy a linguistic privilege. The development of common languages will inevitably prove more advantageous to some groups than to others, although this is something that public policies should try to compensate for. It is far too easy to make a society more egalitarian by making polyglots illiterate, but an enlightened social policy ought to attempt to make the illiterate polyglot.

Multilingualism as a possible solution is strongly argued by a multiculturalist such as May (2003) and by a cosmopolitan such as Van Parjis (2005). As far as education is concerned, and irrespective of the polemical fervor that has fuelled the debate so far, it seems possible to say that cosmopolitans want education to be carried out in the language of the majority for everybody and that the language of the minority be taught as a second language, whereas multiculturalists desire the opposite: that is, they want the principal language be that of each community and the dominant language to be taught as a second language to the minority. See Patten (2003a). 14 See Baker and Prys Jones (1998). The case is argued on the multiculturalist front, by May (2003).


References Anderson, B. (1991) Imagined Communities. London: Verso. Annamalai, E. (2001) Managing Multilingualism in India. Political and Linguistic Manifestations. London: Sage. Barber, B. (1984) Strong Democracy. Participatory Politics for a New Age. Berkeley CA: University of California Press. Baker C. and S. Prys Jones (eds.). (1998) Encyclopaedia of Bilingualism and Bilingual Education. Clevedon PA: Multilingual Matters. Beetham, D. (2000) Democracy and Human Rights. Cambridge: Polity Press. Castiglione, D. and Longman, C. (eds.). (2005) The Challenge of Multilingualism in Law and Politics. Oxford: Hart Publishing. Chambers, C. (2003) “Nation-building, Neutrality and Ethnocultural Justice: Kymlicka's ‘liberal pluralism,’” Ethnicities 3: 295-319. Chandholke, N. (2005) “Negotiating Linguistic Diversity in Democracies: A Comparative Study of the U.S.A. and India,” in S. Bajpeyi (ed.), Democracy and Diversity: A Comparative Study of India and the U.S., forthcoming. Crystal, D. (2000) Language Death. Cambridge: Cambridge University Press Cloots, A. (1793) La République Universelle, in A. Clootes, Œuvres. Paris: Kraus Reprint, 1980. Crawford, J. (2000) At War with Diversity: U.S. Language Policy in an Age of Anxiety. Buffalo NY: Multilingual Matters. Dryzek, J. (2000) Deliberative Democracy and Beyond: Liberals, Critics, Contestations. New York: Oxford University Press. Eco, U. (1995) In search for perfect language. Oxford: Blackwell, 1995. Flaubert, G. (1869) L’education sentimentale. Paris: Bordas, 1974. Glasius, M., Kaldor, M., and Anheier, H. (2001, 2002, 2003) Global Civil Society Yearbook. Oxford: Oxford University Press. Habermas, J. (1998) The Inclusion of the Other. Studies in Political Theory. Cambridge MA: MIT Press. Huntington, S. (2004) “The Hispanic Challenge,” Foreign Policy, March-April: 30-45. Kaldor, M. (2003) Global Civil Society. An Answer to War. Cambridge: Polity Press. Keane, J. (2003) Global Civil Society? Cambridge: Cambridge University Press. Kelly, P. (ed.) (2002) Multiculturalism Reconsidered. Cambridge: Polity Press. Kohli, A. (ed.) (2000). The Success of Indian Democracy. Cambridge: Cambridge University Press. Kymlicka, W. (1995) Multicultural Citizenship. Oxford: Oxford University Press. Kymlicka, W. (2001) Politics in the Vernacular. Oxford: Oxford University Press. Kymlicka, W. and Patten, A. (eds.) (2003) Language Rights and Political Theory. Oxford: Oxford University Press. Mamadouh, V. (2002) “Dealing with Multilingualism in the European Union: Cultural Theory Rationalities and Language Policies,” Journal of Comparative Policy Analysis 4: 327-345. May, S. (2003) “Misconceiving Minority Language Rights: Implications for Liberal Political Theory,” in W. Kymlicka and A. Patten (eds.), Political Theory and Language Policy. Oxford: Oxford University Press. McRae, K. (1983) Conflict and Compromise in Multilingual Societies: Switzerland. Waterloo, Ontario: Wilfrid Laurier University Press. McRae, K. (1986) Conflict and Compromise in Multilingual Societies: Belgium. Waterloo, Ontario: Wilfrid Laurier University Press. Nettle, D. and Romaine, S. (2000) Vanishing Voices: The Extinction of the World’s Languages. Oxford: Oxford University Press. Parekh, B. (2002) Rethinking Multiculturalism. Cultural Diversity and Political Theory. Cambridge MA: Harvard University Press. Patten, A. (2003a) “What Kind of Bilingualism?” in W. Kymlicka and A. Patten (eds.), Political Theory and Language Policy. Oxford: Oxford University Press. Patten, A. (2003b) “Liberal Neutrality and Language Policy,” Philosophy and Public Affairs 31: 356-386. Phillipson, R. (2003) English-Only Europe? Challenging Language Policy. London: Routledge. Pogge, T. (2003) “Accomodation Rights for Hispanics in the United States,” in W. Kymlicka and A. Patten (eds.), Political Theory and Language Policy. Oxford: Oxford University Press.

Van Parijs, P. (2003) “Linguistic Justice,” in W. Kymlicka and A. Patten (eds.), Political Theory and Language Policy. Oxford: Oxford University Press. Van Parijs, P. (2005) “Europe’s Three Language Problems,” in D. Castiglione and C. Longman (eds.) The Challenge of Multilingualism in Law and Politics. Oxford: Hart Publishing. Young, I. M. (2000) Inclusion and Democracy. Oxford: Oxford University Press. Zamenhof, L. L. (1889) An Attempt Towards an International Language, translated by Henry Phillips Jr. New York: H. Holt.


Sign up to vote on this title
UsefulNot useful